Balanced AES/EBU (SPDIF) input circuit (to DIR9001) - Page 2

Bill Fuss · 23rd May 2010, 06:10 PM

The DIR9001 won't know the difference.

novec · 23rd May 2010, 06:29 PM

The DIR9001 accepts both S/PDIF and AES/EBU, so AFAIK the only thing I need is level matching. I imagine it would be a pretty simple thing for an experienced electronics dude to cook up, but I'm not one of them...

The cables will go straight from the soundcard to the DIR9001 without any connectors, so off-the-shelf solutions won't do much good.

Bill Fuss · 23rd May 2010, 07:42 PM

Your Lynx manual says it outputs AES3 on the XLRs. Terminate it into a 110 ohm resistor, couple one end to ground through a .01 cap and couple the other end through another .01 cap to the input pin. Somewhere one line has to terminate into ground. The signal is differentially balanced and the DIR is not. A pulse trafo termination is still the best way though.

stokessd · 08:49 PM

At the risk of sounding like a broken record, i'd look at the Crystal receiver data sheet. The problem with that data sheet is that the receiver portion assumes a single input into the 8416 input MUX. When you have multiple inputs, chances are that you can't use a transformer for each as the data sheet shows. For example I've got both TOSLink and SPDIF going into the receiver chip. So this chip is also effectively single ended because I can't take the receiver negative only for the SPDIF input.

My solution was to capacitively couple all inputs and the return to ground. That works as the Crystal datasheet says for the TOSLink input, but that can't work for a transformer in the SPDIF line. If I wanted a direct connection with the SPDIF, I'd drop a 75ohm resistor to ground, and couple to the chip through a 0.01uf cap and I'd be done. But I wanted to use an input transformer so I did something like this. The transformer primary is floating on both sides and goes to the BNC connector (you would use an XLR). One leg of the secondary is attached to digital ground and the other leg goes to the receiver chip through a 0.01uF cap and also to ground through a 75ohm resistor. I have the benefit of isolation and also input switching with this scheme. I'm not sure if it's ideal, but it has been working well for me.

here's a schematic from anther thread that shows what I did, and how when using the 8416 input MUX, that the input is really also single ended.

Maybe I'm missing something, but I do't see why that wouldn't work for the DIR9001, which is also single ended.
Sheldon

rfbrw · 23rd May 2010, 09:16 PM

Quote:

Originally Posted by Bill Fuss

The DIR9001 won't know the difference.

No, it probably won't. Not after it has been hit by the 7Vpp from the AES16.

rfbrw · 23rd May 2010, 09:28 PM

Quote:

Originally Posted by novec

The DIR9001 accepts both S/PDIF and AES/EBU, so AFAIK the only thing I need is level matching.

Since you are going to have to pad the voltage to SPDIF/TTL levels, you might as well use a tx coupled balanced input chip, a setup that likely mirrors the inputs to the AES16.

novec · 23rd May 2010, 10:01 PM

Is there any reason I can't just terminate negative to ground with a 56 Ohm resistor and put a pad on the positive? I'm not entirely sure why, but I'd prefer to avoid the transformer, unless that's the only passive way to do it. And any kind of chip requires yet another power supply, making it more complicated again. The life of a DIY-er is never easy ;-)

If transformer is the best/only option, do you have any relatively affordable suggestions? Preferrably in the Newark/Farnell inventory.

Bill Fuss · 24th May 2010, 02:09 AM

Properly terminated into 110 ohms you will have a 3.5VP-P signal at the input if the AES16 output is 7VP-P. You don't need a pad, the termination pads it 6db.

Again, use a 110 ohm resistor across the lines and couple the lines to the input pin and ground with .01 caps. It's simple

rfbrw · 24th May 2010, 04:25 AM

Quote:

Originally Posted by novec

Is there any reason I can't just terminate negative to ground with a 56 Ohm resistor and put a pad on the positive? I'm not entirely sure why, but I'd prefer to avoid the transformer, unless that's the only passive way to do it. And any kind of chip requires yet another power supply, making it more complicated again. The life of a DIY-er is never easy ;-)

If transformer is the best/only option, do you have any relatively affordable suggestions? Preferrably in the Newark/Farnell inventory.

Will 4 resistors do? Fig 1.

novec · 24th May 2010, 12:15 PM

Guess what I just found? My brain! I had left it in the fridge for safe keeping friday night, but it's refreshed and ready for action now ;-)

Thanks!

23rd May 2010, 08:45 PM	#14
stokessd diyAudio Member Join Date: Nov 2002 Location: Fort Wayne, IN	At the risk of sounding like a broken record, i'd look at the Crystal receiver data sheet. The problem with that data sheet is that the receiver portion assumes a single input into the 8416 input MUX. When you have multiple inputs, chances are that you can't use a transformer for each as the data sheet shows. For example I've got both TOSLink and SPDIF going into the receiver chip. So this chip is also effectively single ended because I can't take the receiver negative only for the SPDIF input. My solution was to capacitively couple all inputs and the return to ground. That works as the Crystal datasheet says for the TOSLink input, but that can't work for a transformer in the SPDIF line. If I wanted a direct connection with the SPDIF, I'd drop a 75ohm resistor to ground, and couple to the chip through a 0.01uf cap and I'd be done. But I wanted to use an input transformer so I did something like this. The transformer primary is floating on both sides and goes to the BNC connector (you would use an XLR). One leg of the secondary is attached to digital ground and the other leg goes to the receiver chip through a 0.01uF cap and also to ground through a 75ohm resistor. I have the benefit of isolation and also input switching with this scheme. I'm not sure if it's ideal, but it has been working well for me. here's a schematic from anther thread that shows what I did, and how when using the 8416 input MUX, that the input is really also single ended. Maybe I'm missing something, but I do't see why that wouldn't work for the DIR9001, which is also single ended. Sheldon Last edited by stokessd; 23rd May 2010 at 08:49 PM.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Digital SPDIF-AES/EBU interconnect	Nuuk	Digital Source	3	11th April 2012 07:22 PM
Pulse transformers SPDIF/AES-EBU	Kurt von Kubik	Digital Line Level	14	14th November 2009 04:50 PM
AES/EBU to SPDIF coax input, how?	tubenut	Digital Source	48	16th February 2005 07:45 AM
SPDIF to AES/EBU	Kari	Digital Source	4	21st February 2004 11:29 PM
DIGI balanced AES/EBU -> SPDIF Coax	meta	Digital Source	3	26th May 2003 05:51 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

23rd May 2010, 06:10 PM	#11
Bill Fuss RIP Join Date: Oct 2005 Location: Central Florida	The DIR9001 won't know the difference.

23rd May 2010, 06:29 PM	#12
novec diyAudio Member Join Date: Aug 2005	The DIR9001 accepts both S/PDIF and AES/EBU, so AFAIK the only thing I need is level matching. I imagine it would be a pretty simple thing for an experienced electronics dude to cook up, but I'm not one of them... The cables will go straight from the soundcard to the DIR9001 without any connectors, so off-the-shelf solutions won't do much good.

23rd May 2010, 07:42 PM	#13
Bill Fuss RIP Join Date: Oct 2005 Location: Central Florida	Your Lynx manual says it outputs AES3 on the XLRs. Terminate it into a 110 ohm resistor, couple one end to ground through a .01 cap and couple the other end through another .01 cap to the input pin. Somewhere one line has to terminate into ground. The signal is differentially balanced and the DIR is not. A pulse trafo termination is still the best way though.

23rd May 2010, 10:01 PM	#17
novec diyAudio Member Join Date: Aug 2005	Is there any reason I can't just terminate negative to ground with a 56 Ohm resistor and put a pad on the positive? I'm not entirely sure why, but I'd prefer to avoid the transformer, unless that's the only passive way to do it. And any kind of chip requires yet another power supply, making it more complicated again. The life of a DIY-er is never easy ;-) If transformer is the best/only option, do you have any relatively affordable suggestions? Preferrably in the Newark/Farnell inventory.

24th May 2010, 02:09 AM	#18
Bill Fuss RIP Join Date: Oct 2005 Location: Central Florida	Properly terminated into 110 ohms you will have a 3.5VP-P signal at the input if the AES16 output is 7VP-P. You don't need a pad, the termination pads it 6db. Again, use a 110 ohm resistor across the lines and couple the lines to the input pin and ground with .01 caps. It's simple

24th May 2010, 12:15 PM	#20
novec diyAudio Member Join Date: Aug 2005	Guess what I just found? My brain! I had left it in the fridge for safe keeping friday night, but it's refreshed and ready for action now ;-) Thanks!